Use the Console Operator
Every Console Operator task, from your first request to schedules, gates, agents, and connections.
This guide is organized by task. Find what you want to do in the list below, then follow the steps.
The Console Operator is the chat surface in Console. Describe your goal, and it plans the work, hands each part to the right agent, and returns the result in the same conversation. It works inside the selected project.
Ask the Operator for something
Start from a ready-made request
Go back to an earlier conversation
Start over
Attach a file to my request
Choose the model the Operator runs on
Choose how hard the Operator thinks
See how full the conversation is
Follow what the Operator and its agents are doing
Answer a question an agent asks
Keep going when a task runs long
Find what the conversation built
Build an app, agent, or knowledge base
Work with documents
Create or change a design system
Ask about the account
Stop the Operator using a system agent
Let the Operator use one of my agents
Let the Operator act in a connected system
Connect a new system
Run the Operator on a schedule
Pause, edit, or delete a schedule
See conversations waiting for me
Work in a different project
Start
Start a conversation
Select Operator in the Console sidebar. The home greets you by name, with Ask a question or tell me what you need, I'll bring the outcome. Type your request in How can I help you today? and send it.

Use a starter
Pick a starter from the row under the composer, such as Approval Flow or Review a Policy. The arrows either side scroll through them.
Find an earlier conversation
On the home, Recent chats lists your latest conversations. View all chats opens the full list.
In a conversation, Conversation History on the left lists your conversations in the project, newest first. Conversations are specific to each user. Type in Search conversations to filter them by title.
Start a new conversation
Select + New Conversation, top right.
The composer
The composer at the bottom holds your message and four controls. Your model and effort choice applies to each message you send, so you can switch mid-conversation. The composer remembers your last choice for your next conversation.
Attach a file
Select +, then Attach files.
Choose the model
Select the model control, then a model, such as Gemini 3.7 Flash or GPT 5.4.
Choose the effort
Select the effort control, then Minimal, Low, Medium, or High. More effort usually gives better plans on complex requests, at the cost of time and credits.
See how full the conversation is
The context gauge shows how full the conversation's context is, such as 0%. Hover over it for the detail, such as 18 / ~200.0K tokens. When it fills up, start a new conversation.
Work in a conversation
A conversation has three columns: Conversation History on the left, the transcript in the middle, and the side panel on the right.

Follow the work
The transcript shows how long the Operator thought, each step it took, such as Update Plan Steps, and each question it asked with the answer beneath it.
When it hands work to an agent, the agent appears as a card with its step count, such as App Builder Agent · 49 steps, and a Working status while it runs.
Select Open full run on the card to see every step the agent took.
Select Jump to latest to return to the bottom.
Answer a question from an agent
When an agent needs a decision, the question appears as a card above the composer. Choose one:
Pick one of the numbered answers.
Write your own in Answer in your own words….
Attach a file, when the agent asks for one, such as a brand guide.
Select Skip to let the agent carry on without an answer.

A conversation that ends on a question also appears on Gates until you answer.
Keep going when a task runs long
When a turn runs long, This is taking a while appears above the composer. Choose one:
Continue in a new conversation — carry on elsewhere, with a summary of where things stand.
Stay here — keep watching.
The work keeps running whichever you choose.
Find what the conversation built
In the side panel:
Outputs — what the conversation built, such as an app, with a link to open it.
Artifacts — files and documents created during the task.
If the side panel is hidden, select the panel icon beside + New Conversation.
What you can ask for
Build an app, agent, or knowledge base
Ask for it, and the Operator creates it in the current project. You can test a new agent in the same conversation.
An app — the App Builder Agent asks which of the project's design systems to use, or name one in your request.
Knowledge from a website — give it a URL, and it can use the site's content to create a knowledge base.
Building an app or ingesting knowledge can take several minutes. The Operator reports the real outcome when the build finishes, even if the conversation was interrupted.
Work with documents
Ask the Document AI Agent to extract data from documents, classify or summarise them, answer questions from them, or generate new ones. Before it generates a document, it lists the project's saved document templates and asks you to choose one.
Create or change a design system
Ask the Operator to create, change, or delete one of the project's design systems, for example from a brand's website. To set a logo, attach the image to the conversation. See Create a design system.
Ask about the account
Ask who is in the account and which teams it has. The Operator can tell you, but it can't add, remove, or change people or teams. That stays in Users and Teams. It can also manage the account's projects for you.
Agents and connections
The side panel's Agents and Connections sections decide what the Operator may use.
Stop the Operator using a system agent
Under Agents › System, select the agent's ⋯, then turn it off. System agents start on.
The setting applies to you, in this project, and sticks across new conversations. A change takes effect on your next request. You can't turn an agent off while it is working on a run.
Let the Operator use one of my agents
Under Agents › User, select the agent's ⋯, then turn it on. Your own agents start off. Only agents on the current agent runtime are listed.
Let the Operator use a connection
Under Connections, select the connection's ⋯, then Enable. Every connection starts off. You can also select + › Connections in the composer.
The ⋯ menu also has:
Manage — choose which of the connection's actions the Operator may use, and how closely each is supervised.
Manage in Integrations — open the connection's details in the integrations catalog.
Only the Operator itself uses these connections. Agents it hands work to don't inherit them.
Connect a new system
Under Connections, select All integrations. Connect one in the catalog. It appears in Connections when you close the window, turned off until you turn it on.
For the system agents and how each setting behaves, see Agents & connections.
Schedules and gates
Run the Operator on a schedule

A scheduled run acts with no one watching, and no one can answer its questions. Give it everything it needs in the instructions.
You can also describe the schedule in a conversation, for example "Every Monday at 8 AM UTC, check which knowledge bases have files that need syncing and tell me." The Operator sets it up for you.
Pause, edit, or delete a schedule
On Schedules, select the schedule's ⋯, then Pause, Resume, Edit, or Delete. Type in Search schedules… to find one. There is no Run now; to run the same instructions straight away, ask the Operator.
See conversations waiting for me
Select Gates in the top bar. It lists the conversations you started that have stopped to ask you something, each with its title, the agent asking, and how long it has waited. Select one to open the conversation at the question, then answer it.

Only you can answer your requests. Once you answer, the request leaves Gates and the work continues.
Work in a different project
Open the account and project switcher at the top of the Console sidebar, then choose a project. Everything the Operator creates belongs to the selected project, so switching changes what it can reach.

Common issues
I can't see a colleague's conversation. Conversations are specific to each user; you see only the ones you started.
The Operator didn't use my agent. Your own agents start off. Turn it on under Agents › User. Agents on an older runtime aren't listed.
The Operator says it has no connection for a system. Turn on the connection it names, or connect one from All integrations.
Something I started hasn't finished. Check Gates: it may be waiting for your answer.
The context gauge is nearly full. Start a new conversation.
A scheduled run didn't do what I expected. Nobody can answer its questions. Put everything it needs in the instructions, and limit Agents this run may use to what the task needs.
Related pages
Maintained by Abhishek Paul
AI-assisted, human-approved

Last updated
Was this helpful?

